Inside a plain white cylinder a bit bigger than a beer keg at the University of Sydney is one of the coldest places in the universe.
It sits at about a hundredth of a degree above absolute zero.
Within the cylinder is a piece of technology forged in a silicon furnace, which aims to revolutionise everything from medicine, to our understanding of climate change, and even banking.
The world’s biggest tech companies including Microsoft, IBM and Google are spending billions of dollars on experiments like this.
Start-ups are popping up all over the globe, hoping to cash in on the technology. Despite the fact it is still being invented, scientists are already writing the apps that will run on the tech.
“The world is hungry for computing power,” said quantum physicist David Reilly, from the University of Sydney.
The centre he heads up, Microsoft Station Q Sydney, aims to do one key thing: develop a useful, scalable quantum computer.
A quantum computer is one that encodes the basic units of information — the quantum version of a “bit”, known as “qubit” — using the spooky properties held by quantum particles, rather than in the transistors used in traditional computers.
“We’ve been grappling with the basic physics and the meaning of quantum mechanics,” Professor Reilly said.
Despite that grappling, he and many others like him are using that cutting-edge physics to build an entirely new type of computer.
Quantum mechanics reveals matter is not what it seems.
For example, it reveals some objects — small particles like electrons, atoms and even molecules — can be in two places at the same time.
They can have two opposite properties simultaneously too, like having both a positive and a negative charge.
Weird facts like that are now being used by scientists, engineers and software developers to create machines set to change the world in ways that are literally unimaginable.
“It’s exploiting laws of nature that we’re only just understanding,” Professor Reilly said.
“We can already see that if we had a true, universal quantum computer we could attack some of the world’s grandest challenges.
“Areas of health, energy, climate change — those types of problems require enormous computing power. And we think these different ways of computing will be useful and important.”
The quantum Silicon Valley
Whether through accident or design, Australia — and Sydney in particular — has become a world leader in the push to make quantum computers a reality.
“The Sydney area now, I think, has approximately the highest concentration of quantum scientists and engineers of any city in the world,” Professor Andrew Dzurak said.
He is based at the University of New South Wales (UNSW) and leads another Sydney-based group pushing the quantum boundaries.
UNSW is joining forces with the University of Sydney, Macquarie University, the University of Technology and a string of start-ups and tech giants, with the hopes of it becoming the place to be for quantum tech.
A funding announcement by the NSW Government could push Sydney over that verge, with the establishment of a Sydney Quantum Academy.
“We’re announcing a $15.4 million investment in Sydney’s quantum future,” NSW Minister for Innovation Matt Kean said.
The Government believes that money will be matched by industry and universities to create a $35 million investment in the technology over five years.
The academy will provide a pathway for quantum scientists, engineers and software developers to train and then work in Sydney.
It will provide PhD and post-doctoral positions that allow students to work across all of the institutions partnering in the academy, helping to foster collaboration among the various groups that sometimes engage in fierce competition.
The Sydney Quantum Academy will also aim to incubate the start-ups and ‘spin-off companies’ that will employ those students and attract the massive investment from the private sector needed to turn quantum computers into a commercial reality.
Professor Reilly said training people with the unique expertise required to work on quantum computers was the key to turning Sydney into a quantum powerhouse.
“The thing that’s missing, and it hits me in the face every single day, is people,” he said.
“In order to understand the theoretical underpinnings of quantum mechanics and see how to translate that in an engineering setting requires a set of skills [and] we don’t have too many people that possess all those skills today.”
Professor Dzurak thinks with the right support, Sydney can become the Silicon Valley of quantum computing.
“Silicon Valley is successful because it has lots of companies, lots of expertise, all centralised in one area. And that’s what we’ve already got the genesis of here. And I think with the Sydney Quantum Academy, we can really see this realise its potential,” he said.
“I think we’ve got the potential, really, to create a quantum Silicon Valley here in the Sydney region.”
So how will the quantum computers of the future work?
Imagine you have a mathematical problem and a series of potential solutions.
If you ask a traditional digital computer which is the correct one, it will need to check each potential solution in a series: one after the other. The longer the list, the longer the time it will take.
But for at least some types of problems, quantum computers will be able to check all potential solutions at once.
In addition, with traditional digital computers, to double the speed at which it operates, you need to double its number of bits.
But with quantum computers, the speed they can operate at doubles every time you add a new qubit.
Those two properties combined means the moment quantum computers outpace traditional computers, they are on an exponential path to unimagined computing power.
Take internet cryptography as an example. The fact that traditional computers work in series rather than parallel is relied on for a huge amount of cryptography.
That sort of cryptography keeps all your banking and sensitive communication safe.
Breaking cryptography with current computers is technically possible — it would just take thousands of years.
It involves a problem for which there are many solutions that traditional computers must examine one after another.
But quantum computers will be able to break cryptography almost instantly by examining all the possible solutions at once.
Even though proper quantum computers don’t exist yet, Sydney has a world-leading team of software engineers coding the apps that will run on them.
Professor Michael Bremner from UTS leads that team.
“You can think of a quantum computer as being a bit like a graphics card on your computer,” Professor Bremner said.
Graphics cards are particularly good at some sorts of problems, and the software decides which parts of a big problem should be sent to the graphics card, and which to throw at the main processor.
“You can think of a quantum processor in a similar way — we’re trying to work out which parts go where.”
When can consumers buy one?
Ask a bunch of quantum computing researchers how far away we are from seeing one and they will all say something different.
Limited sorts of quantum processors already exist.
Professor Reilly said it would only be a few years before we started to see those machines performing tasks we could not do, or do as well, with traditional computers: “A machine that’s been custom-designed to attack a particular problem.”
But the real game is to produce a large “universal quantum computer” — one that can run any quantum app and can do it faster than any traditional computer.
“A decade from now, we’re going to start to see more universal, more generic computing power being captured by quantum computing,” Professor Reilly said.
All the scientists agree the technology will be revolutionary. And they all point to drug design as the most obvious world-changing application of them, since quantum computers will be able to predict the effects of drugs, without the expensive, time-consuming process of trial and error that is necessary today.
But beyond that, and a few other specific things, Professor Reilly said it was hard to know exactly how the technology would change the world.
“The truth is, we don’t have a machine yet. So it’s a little bit challenging, a little bit unfair to say ‘what’s the thing good for’ when you haven’t built it,” he said.
“And I think history teaches us that if you look at the classical conventional computers of the 1940s and 1950s, it was very hard with those machines to anticipate how they’d be used even 10 years, 20 years ahead of time — what the applications would be.
“You kind of have to build it first, learn what it does and exploit it. Then the really big applications will emerge.”